Influence of the Particle Number on Mixture Quality

The aim of the research was to establish regression relationships establishing the interrelation between the number of measurement units in the samples and the mixture quality variability index (coefficient of variation of the control component content in the samples) for the same mixture. As a result of simulation of the obtained models for comparable parts of the particle number, we received practical coincidence of numerical values of different models in the interval of 10…30 particles in the sample, with the greatest coincidence of the models around 20…25 pc. Decreasing the number of particles significantly increases the scatter of variation coefficient values. Increasing the number of controlled components increases the variation of the obtained values. Increasing the size of the particles increases the values of the coefficient of variation. For example, for lentils at 60 (pcs./sample), the coefficient of variation is 15.5%, for barley—10.7% and for buckwheat—9.9%. That is, the differences in the two small components are insignificant and comparable to the error of the experiments. The estimated number of particles in the samples taken is at least 19 pcs. Reducing the calculated value to 14 pcs. (lentils) increases the variability of the index.